While microwave imaging satellites, such as the NASA Soil Moisture Active Passive (SMAP), can provide reliable estimates of surface soil moisture at km resolution, the temporal frequency of observations is on the order of days. To increase the temporal frequency of observations, a new class of approaches considers global navigation satellite system (GNSS)-reflectometry (GNSS-R) signals. In this work, we consider observations from the NASA Cyclone GNSS (CYGNSS) constellation, as well as auxiliary observations, and seek to provide instantaneous soil moisture estimates. To achieve accurate retrievals, a novel machine learning approach for probabilistic regression is considered, namely the NGBoost. In addition to achieving an accuracy comparable to previous approaches employing state-of-the-art machine learning methods, the considered framework also provides prediction intervals to quantify prediction uncertainty. Using observations from the Yanco SMAP core validation site in southeast Australia over a period of three years, we quantify the performance in terms of both retrieval accuracy and associated uncertainty. Furthermore, using noisy observations, we experimentally demonstrate the impact of input noise on the prediction uncertainty.

Monitoring soil moisture enables detailed understandings of its role in the water cycle and how it is affected by climate change. Multiple airborne and spaceborne missions have been dedicated to estimating soil moisture, including Soil Moisture Active Passive (SMAP) [1] , Soil Moisture and Ocean Salinity (SMOS) [2] , and Airborne Microwave Observatory of Subcanopy and Subsurface (AirMOSS) [3] . Some other remote sensing missions have added soil moisture retrievals along with their primary goal. For instance, the primary objective of the Cyclone Global Navigation Satellite System (CYGNSS) mission [4] is wind speed retrieval over the ocean, but it is now also used for soil moisture retrieval, among other applications [5] , [6] . All these remote sensing systems and future systems need in situ soil moisture measurements to validate their products.

A ground-based distributed sensing network is described in this work that leverages elements of wireless sensor networks (WSN) and uncrewed areal vehicles (UAVs) with software-defined radar payloads. Hardware and software advancements are made towards combining the operations of WSNs and UAVs for dynamic spatiotemporal monitoring of surface to subsurface soil moisture at kilometer scales. The multi-agent and distributed sensing approach demonstrates coordination, collaboration, and parallel operation of discrete assets for optimal soil moisture monitoring. Results from the first field experiment showing this coordinated operation are reported.

In this paper we develop a Convolutional Long Short-term memory (ConvLSTM) model, a time series deep learning neural network, to predict soil moisture, with an add-on module of passive microwave (radiometer) soil moisture retrieval using the Tau omega model. We incorporate antecedent observations, landscape properties, and forcing factors such as precipitation, landcover, clay fraction, and brightness temperature in the prediction scheme. A regularization Monte Carlo Dropout layer is added to the network to remove stochasticity and avoid overfitting during the training phase. This dropout layer also provides a Bayesian approximation to quantify uncertainty during forecasting. The model is validated at four Soil Moisture Active Passive (SMAP) Cal/Val locations using performance metrics such as Root Mean Square Error (RMSE) and Bias to evaluate effectiveness of the proposed method. This model is developed as a component of the Science Simulator within the Distributed Spacecraft with Heuristic Intelligence to Enable Logistical Decisions (D-SHIELD) project.

A new observing strategy for floods was demonstrated and evaluated in a testbed environment. The strategy coordinates several observing platforms, including in situ and space based, to observe a flood from multiple vantage points and dynamically target predicted flood events with highresolution observations. The coordinated observations were assimilated back into the model to continuously improve forecasts and future observation selection. The demonstration shows the potential for coordinated, model-driven observing strategies and the feasibility of the NOS Testbed for demonstrating and evaluating new observing strategies.

Adaptive and targeted allocation of mobile sensing agents, in the form of unmanned aerial vehicles (UAVs) with software defined radar (UAV-SDRadar) payloads, enable mapping of surface soil moisture in regions wherein situwireless sensor networks (WSNs) undersample soil moisture or upscaling models perform poorly. This work presents an optimization-based UAV path planning methodology that seeks to maximize UAV flight coverage over areas where a complementing WSN yields upscaled soil moisture estimates with high uncertainty. By recursively mapping soil moisture over such areas, the combined UAV and WSN instrumentation can gradually capture the domain’s true mean soil moisture. A series of numerical simulations are presented to demonstrate the algorithm’s basic function while considering real-world and feasible operational scenarios.

A comparison of three different electromagnetic scattering models for land surface delay-Doppler maps (DDMs) obtained from global navigation satellite system reflectometry (GNSS-R) along a Cyclone Global Navigation Satellite System (CYGNSS) track in the San Luis Valley, Colorado, USA, is presented. The three models are the analytical Kirchhoff solutions (AKS), the Soil And VEgetation Reflection Simulator (SAVERS), and the improved geometrical optics with topography (IGOT). Common inputs to the three models were defined by using field samples of soil moisture and texture, soil surface roughness measurements, and a digital elevation model (DEM). The resulting peak reflectivity profiles of the models and the CYGNSS data all had a range of 10 dB along the selected track, mainly due to the influence of topography. The reflectivities obtained from all three models agreed with one another to within 2.4 dB along the full length of the track. The models also showed general agreement with the corresponding CYGNSS data, although the modeled profiles were higher than CYGNSS Science Data Record Version 3.1 by an average of 5 dB and also smoother. Additional characterization of fine-scale surface roughness is identified as an area for future work to improve model fidelity. An intercomparison of DDM structure for three selected acquisitions is also provided.

A comparison of three different electromagnetic scattering models for delay-Doppler maps (DDMs) of global navigation satellite system reflectometry (GNSS-R) from land is performed along a Cyclone Global Navigation Satellite System (CYGNSS) track over a validation site in the San Luis Valley, Colorado, USA. The peak reflectivity profiles of all three models and of the corresponding CYGNSS data are found to be in general agreement and are strongly influenced by topography. An intercomparison of DDM structure for one acquisition is also included. Efforts to refine the model results using a high resolution lidar survey are ongoing.

We describe a suite of scalable software methods and frameworks to helps schedule payload operations of a large constellation, with multiple payloads per and across spacecraft, such that the collection of observational data and their downlink, constrained by the constellation constraints (orbital mechanics), resources (e.g., power) and subsystems (e.g., attitude control), results in maximum science value for a selected use case. Constellation topology, spacecraft and ground network characteristics can be imported from design tools or existing constellations and can serve as elements of an operations design tool. Our framework includes a science simulator to inform the scheduler of the predictive value of observations or operational decisions. Autonomous, realtime re-scheduling based on past observations needs improved data assimilation methods within the simulator.

An active–passive microwave retrieval algorithm for simultaneous determination of soil surface roughness parameters [vertical root-mean-square (RMS) height (${s}$) and horizontal correlation length (${l}$)] is presented for bare soils. The algorithm is based on active–passive microwave covariation, including the improved Integral Equation Method (I2EM), and is tested with global soil moisture active passive (SMAP) observations. The estimated retrieval results for${s}$and${l}$are overall consistent with values in the literature, indicating the validity of the proposed algorithm. Sensitivity analyses showed that the developed roughness retrieval algorithm is independent of permittivity for${\varepsilon }_{s} > 10$[-]. Furthermore, the physical model basis of this approach (I2EM) allows the application of different autocorrelation functions (ACF), such as Gaussian and exponential ACFs. Global roughness retrieval results confirm bare areas in deserts such as Sahara or Gobi. However, the type of ACF used within roughness parameter estimation is important. Retrieval results for the Gaussian ACF describe a rougher surface than retrieval results for the exponential ACF. No correlations were found between roughness results and the amount of precipitation or the soil texture, which could be due to the coarse spatial resolution of the SMAP data. The extension of this approach to vegetated soils is planned as an add-on study.

This work presents recent field activities in support of the NASA CYGNSS missions' land applications. Land reflected GNSS signals are known to be sensitive to surface topography, vegetation cover, and soil moisture. To better understand CYGNSS sensitivity to surface conditions, especially freeze-thaw states, two SoilSCAPE wireless in situ network sites were deployed in the San Luis Valley (SLV), CO, in late Oct. 2019. These sites capture similar weather and climatic conditions but have contrasting topography and vegetation cover. Initial analysis of CYGNSS Signal-to-Noise (SNR) observations over SLV indicates the need to fully account for land-scape topography. To this end, a forward wave scattering model that incorporates a Digital Elevation Model (DEM)is currently being developed to help explain the effects of local topography on SNR observations.

With the availability of global satellite remote sensing observations of surface soil moisture, it is now possible to quantify important hydrological fluxes such as evapotranspiration (ET) and drainage. Furthermore, given the current level of accuracy of remotely sensed soil moisture, these fluxes can be estimated without the need for large-scale land surface or climate modeling. In this work, remote sensing data from the NASA SMAP mission, at 36 [km] scale, and gauge-based precipitation data over the US are utilized within an adjoint-state variation estimation method to obtain time-series daily estimates of ET and drainage. The approach uses only SMAP measurements and precipitation. Neither a hydrology or land surface model nor ancillary hydrologic data are used. ET estimates are compared to eddy covariance measurements from the AmeriFlux network and are shown to capture up to 70% of the in-situ measurements' annual variance. Similarly, Drainage estimates are compared to USGS streamflow measurements. On average Drainage under-estimates streamflow by 1-2 [mm day-1] with seasonal correlation (R2) varying between 0.52-0.77. These estimates close the surface water budget with only SMAP measurements and precipitation information.

In this paper we describe the development of new wireless sensor network technologies to coordinate among different ground-based and unmanned aerial vehicle (UAV)-based sensors as “Agents” who, when coordinated, deliver ground-truth at varying temporal and spatial sampling scales for NASA remote sensing science products, as well as for other potential users that may have different application requirements.

D-SHIELD is a suite of scalable software tools that helps schedule payload operations of a large constellation, with multiple payloads per and across spacecraft, such that the collection of observational data and their downlink, constrained by the constellation constraints (orbital mechanics), resources (e.g., power) and subsystems (e.g., attitude control), results in maximum science value for a selected use case. Constellation topology, spacecraft and ground network characteristics can be imported from design tools or existing constellations and can serve as elements of an operations design tool. D-SHIELD will include a science simulator to inform the scheduler of the predictive value of observations or operational decisions. Autonomous, realtime re-scheduling based on past observations needs improved data assimilation methods within the simulator.

We report on recent technology advancements and developments in in situ soil moisture wireless sensor networks (WSN) in support of Earth science microwave remote sensing missions. Specifically, we discuss new hardware features, known as Wakeup-on-Radio (WoR), that enable sensor networks to respond rapidly to short-lived micrometeorological events and record measurements which may typically be lost in-between sampling periods. Additionally, we outline strategies towards WSN autonomy with the goal of enabling an in-situ sensor to "learn" soil moisture dynamics and surrounding ecohydrological processes, to then determine its optimum sampling schedule.

Soil roughness strongly influences processes like erosion, infiltration, moisture and evaporation of soils as well as growth of agricultural plants. An approach to soil roughness based on active-passive microwave covariation is proposed in order to simultaneously retrieve the vertical RMS height (s) and horizontal correlation length (l) of soil surfaces from simultaneously measured radar and radiometer microwave signatures. The approach is based on a retrieval algorithm for active-passive covariation including the improved Integral Equation Method (I2EM). It is tested with the global active-passive microwave observations of NASA's Soil Moisture Active Passive (SMAP) mission. The developed roughness retrieval algorithm shows independence of permittivity for εs> 10 [-] due to the covariation formalism. Results reveal that s and l can be estimated simultaneously by the proposed approach since surface patterns of nonvegetated areas can be assessed on global scale. In regions with sandy deserts, like the Sahara or the outback in Australia, determined s and l confirm rather smooth to semi-rough surface roughness patterns with most frequent vertical RMS heights smaller 3 cm and corresponding higher horizontal correlation lengths (> 8 cm).

Active and passive low-frequency microwave measurements from a number of space- and airborne instruments are used to estimate soil moisture. Each of the sensing approaches has distinct advantages and disadvantages. There is increasing interest in combining active and passive measurements in order to realize the advantages and alleviate the disadvantages. In order to combine active and passive measurements, their covariations with respect to soil moisture need to be known. The covariation is dependent on how the active and passive microwaves interact with vegetation canopy and soil surface. In this paper, we introduce a physics-based model for the covariation of active and passive microwaves over soil surfaces with vegetation cover. The analytical form for a covariation function is derived which depends on the scattering and absorption of microwaves by soil and vegetation with different orientations, structures, and water contents. The main finding is that the covariation function β is related to the roughness and vegetation losses in the two measurements. An increase in soil roughness or in vegetation cover leads to less negative values of β, which is pronounced for dense and moist vegetation. Both the soil and vegetation components introduce a polarization dependence of β that is caused by polarization-induced differences in soil scattering and oriented plant structures. The forward modeled covariations are plotted together with statistically derived covariation estimates from two months of global active and passive L-band observations of the Soil Moisture Active Passive mission. The physically modeled and statistically derived estimates of covariation are comparable in magnitude and scale.

Hydrological water balance studies require knowledge of terrestrial water content within an active storage volume. While the advent of microwave remote sensing has enable frequent and global observations of surface soil moisture, by itself, surface moisture content is not directly suitable for use in water balance studies. An appropriate length-scale, Δz, is required to transform the soil moisture state to total water content. This work demonstrates the applicability of SMAP soil moisture in first-order and observation-driven water balance. First, a novel method is presented to estimate the soil water loss function. Then, by resolving the water balance equation and enforcing mass conservation, estimates of the hydrological length scale over the United States is provided. Mean precipitation is the dominant factor on Δz, such that wetter regions with higher mean precipitation have larger lengths scale. The mean soil moisture state and texture weakly influence Δz.

In this study, the τ-ω model framework is used to derive extinction coefficient and canopy penetration depths from multi-frequency SMAP and AMSR2 retrievals of vegetation optical depth together with ICESat LiDAR vegetation heights. The vegetation extinction coefficient serves as an indicator of how strong absorption and scattering processes within the canopy attenuate microwaves at L and C-band. Through inversion of the extinction coefficient, the penetration depth into the canopy can be obtained, which is analyzed on local (Sahel, Illinois) and continental scale (Africa, parts of North America) as well as for a one year time series (04/2015-04/2016). First analyses of the retrieved penetration depth estimates reveal strongest attenuation for densely forested areas, therefore vegetation attenuation should be accounted for when retrieving soil moisture in these areas. For the continents of North America and Africa penetration depths decrease in average with an increase in frequency from L- to C-band. Moreover penetration depth time series were found to match with expected seasonal variations (e.g. vegetation growth period & rainy season) for analyzed local regions.

In this study we evaluate a Random Forest (RF) model for characterizing the spatial variability of soil moisture based on model derived from in situ soil moisture samples, geophysical data and RADAR observations. The RF model is run with and without C-band SAR backscatter to understand the importance of the inclusion of SAR data for mapping of soil moisture at field scale. The inclusion of SAR data in the RF resulted in a modest improvement however the geophysical parameters (e.g. soil types and terrain properties) were of greater importance.

SMAP and SMOS missions estimate soil moisture using a zeroth-order radiative transfer model, the τ-ω model. Its simplifying assumption of a weakly scattering vegetation medium is insufficient in the presence of woody biomass, greater than 30% of the land surface. Here, a simplified first-order radiative transfer equation for use in retrieval algorithms is proposed. The inclusion of first-order scattering increases sensitivity to soil moisture especially for wet surfaces. The recently developed multi-temporal dual channel algorithm (MT-DCA) is implemented over Africa using both the τ-ω model and the proposed first-order equation with SMAP 36 km gridded brightness temperature measurements as inputs. The algorithm finds large changes in soil moisture mean and standard deviation in areas with woody vegetation and little change elsewhere. This implies that inclusion of first-order scattering can significantly change mean soil moisture retrievals and increase their temporal variability in regions with woody biomass.

Combined active-passive remote sensing has the potential for capturing the relative advantage of each sensing approach in geophysical retrievals. One cornerstone of combined active-passive microwave sensing is the modeling of the covariation of active and passive signals, which arise from equivalent sensitivities of both sensor types to changes in geophysical properties. In this research contribution, we propose a physics-based active-passive combination of active and passive microwave observations based on Kirchhoff's law of energy conservation. This allows establishing a physics-based forward model as well as a fully data-driven, single-pass retrieval methodology for active-passive microwave covariation. The forward model and the retrieval approach are adaptable to different sensor characteristics (incidence angle, frequency & polarization). The theoretical (forward model) as well as applied (retrieval method) physics-based covariation framework is tested with SMAP (LL) and SMAP/Sentinel-1 (LC) data to reveal potentials and constraints for active-passive microwave sensing. As a result of the conducted study, a linear functional relationship between active and passive microwave observations (e.g. assumed for the SMAP mission) is confirmed, if higher-order scattering can be omitted.

Decomposition is performed for the 4×4 SMAP radar channel covariance matrix and the correlation between resulting components, surface soil moisture and vegetation is examined. Globally, the first principal component is the most dominant and the correlation coefficients with respect to soil mortise is highest (R2≥ 0.8) in regions with fractional ground cover and sufficient temporal dynamics of soil moisture.

In this study we examined the linear decomposition and relationship between the SMAP observed Polarization Ratio into surface soil moisture and vegetation. Temporal linear regression, per each SMAP pixel, is performed to estimate the decomposition coefficients. Variances (explained variance) in PR is predominantly dominated by dynamics of surface soil moisture and degrades with increasing vegetation amount. Although PR, by itself, is high in arid and semi-arid regions, due to lack of moisture and vegetation dynamics, the explained variance is very small.

Soil moisture impacts exchanges of water, energy and carbon fluxes between the land surface and the atmosphere. Passive microwave remote sensing at L-band can capture spatial and temporal patterns of soil moisture in the landscape. Both ESA and NASA have launched L-band radiometers, in the form of the SMOS and SMAP satellites respectively, to monitor soil moisture globally, every 3-day at about 40 km resolution. However, their coarse scale restricts the range of applications. While SMAP included an L-band radar to downscale the radiometer soil moisture to 9 km, the radar failed after 3 months and this initial approach is not applicable to developing a consistent long term soil moisture product across the two missions anymore. Existing optical-, radiometer-, and oversampling-based downscaling methods could be an alternative to the radar-based approach for delivering such data. Nevertheless, retrieval of a consistent high resolution soil moisture product remains a challenge, and there has been no comprehensive intercomparison of the alternate approaches. This research undertakes an assessment of the different downscaling approaches using the SMAPEx-4 field campaign data.

The synergy of active and passive microwave measurements have attracted increasing attention in recently years. In this study, we investigate the relationship and covariation of the SMAP radar backscatter and radiometer reflectivity as a function of surface roughness and vegetation. Two radar-derived indices, namely the radar vegetation index (RVI) and radar roughness index (RRI) are adopted to account for the contributions from vegetation and surface roughness respectively. The results show RVI distinguishes vegetation density well in sparse to densely vegetated regions, while significantly overestimates the biomass over some dry desert regions due to possible soil volume scattering effects. RRI well captures the negative covariation of active and passive measurements in bare and sparsely vegetated surfaces, while becomes ineffective in densely vegetated areas due to the reduced contribution from soil surfaces.

A robust physics-based combined radar-radiometer, or Active-Passive, surface soil moisture and roughness estimation methodology is presented. Soil moisture and roughness retrieval is performed via optimization, i.e., minimization, of a joint objective function which constrains similar resolution radar and radiometer observations simultaneously. A data-driven and noise-dependent regularization term has also been developed to automatically regularize and balance corresponding radar and radiometer contributions to achieve optimal soil moisture retrievals. It is shown that in order to compensate for measurement and observation noise, as well as forward model inaccuracies, in combined radar-radiometer estimation surface roughness can be considered a free parameter. Extensive Monte-Carlo numerical simulations and assessment using field data have been performed to both evaluate the algorithm's performance and to demonstrate soil moisture estimation. Unbiased root mean squared errors (RMSE) range from 0.18 to 0.03 cm3/cm3 for two different land cover types of corn and soybean. In summary, in the context of soil moisture retrieval, the importance of consistent forward emission and scattering development is discussed and presented.

With emphasis on physics-based techniques, a multi-objective optimization approach to combined radar-radiometer soil moisture estimation is presented in this work. Soil moisture estimation is demonstrated via application of this method to SMAP high resolution radar and coarse resolution radiometer data. Comparisons are then made with the SMAP baseline active-passive soil moisture output data product. A strong agreement between the two techniques, especially in capturing spatial distributions of soil moisture is observed.

Improved soil moisture estimation, within a combined radar-radiometer framework, by using Joint-physics modeling is demonstrated and presented in this work. Detailed numerical simulation on bare, but rough surfaces, highlight how improved emission-scattering modeling, can outperform conventional approaches. Root-mean-squared error statistics show almost a factor of two improvement in soil permittivity predictions.

In order to provide a reliable source of ground-based validation data for the SMAP mission at spatial scales of 3 km, 9 km and 36 km, we have developed a new regression-based method capable of yielding highly-accurate upscaled soil moisture estimates based on sparse, irregularly-spaced soil moisture measurements.

We present a method to combine same-resolution measurements of active radar and passive radiometer microwave remote sensing to build a framework for soil moisture estimation in support of the Soil Moisture Active and Passive (SMAP) mission. A unified active-passive soil moisture estimation algorithm is developed within a global optimization scheme, using a joint cost function with adaptive regularization, where, unlike traditional methods, both radar and radiometer measurements are utilized at the same time to retrieve soil moisture. Monte Carlo numerical simulations and optimization to retrieve soil moisture are performed for Corn, Soybean, and Grass landcover types for active-only, passive-only, and active-passive scenarios. These numerical experiments show that the proposed combined active-passive (CAP) soil moisture estimation approach outperforms either the single-sensor technique, particularly for higher vegetation water content (VWC) values (VWC > 3 kg/m2). For example, for the case of Corn with VWC of 5 kg/m2, retrieval error is reduced to 0.035 cm3/cm3for the active-passive method from 0.08 cm3/cm3for active scenarios. Furthermore, tests of this new algorithm on the Passive and Active Land S-band Sensor (PALS) Soil Moisture Experiment 2002 (SMEX02), as well as the Combined RadarRadiometer (ComRad) collocated active and passive data, demonstrate the applicability of this method to actual data, even with potentially inaccurate forward models and noisy data. Results indicate that the best soil moisture estimates over a large range of soil moisture (0.04-0.4 cm3/cm3) and vegetation (0-5 kg/m2) conditions are achievable when the adaptive regularization parameter γ is chosen to give slightly more weight to the radiometer forward model without discarding the complementary radar measurement points.

A combined radar-radiometer soil moisture estimation framework is presented in this work, which utilizes both radar backscatter and radiometer brightness temperature using physics-based models that fundamentally couple the scattering and emission processes. A regularization or tuning parameter is introduced within the optimization algorithm to enable flexibility and adaptability. It is observed that by finding the right balance between radar and radiometer contributions within a “joint” cost function, best estimates over a larger range of surface soil moisture and roughness are achievable. Monte Carlo numerical simulations are performed to highlight how this novel Active-Passive method is capable of fully utilizing emission and scattering sensitivities to surface soil moisture.

The ground-based University of Florida L-band Automated Radar System (UF-LARS) was developed to obtain observations of normalized radar backscatter (\mmbσ0) at high temporal resolution for soil moisture applications. The system was mounted on a 25 m manlift with capabilities of antenna positioning for multi-angle data acquisition and ranging. The RF subsystem of UF-LARS was based upon the established designs for ground-based scatterometers employing a vector network analyzer with simultaneous acquisition of V- and H-polarized returns. System integration and automated data acquisition were enabled using a software control system. Fifteen-minute observations of \mmb σ0collected over a growing season of sweet-corn and bare soil conditions in North Central Florida, were used to study the sensitivity of \mmbσ0to growing vegetation and near-surface (0-5 cm) soil moisture (\mmbSM0 - 5). On average, \mmb σ\mmbVV0were observed to be 23% higher than \mmbσ\mmbHH0during the mid- and late-stages of crop growth due to the vertical structure of stems. The correlation between 3-day observations of \mmbSM0 - 5 and \mmbσ\mmbVV0reduced by 55% compared to those obtained for ≤ 30-min observations. These findings suggested that data set at high temporal frequencies can be used to develop more realistic and robust forward backscattering models.

A soil moisture retrieval algorithm is presented whereby both radar and radiometer measurements are used simultaneously in an optimization scheme to retrieve for surface soil moisture in the presence of vegetation. Further, using fine-resolution radar measurements, a disaggregated brightness temperature product at the radar resolution is developed to reconcile the spatial resolution discrepancy between the two measurements. Numerical simulations are performed to synthesize SMAP data, and then via the method of Simulated Annealing, optimization is accomplished to retrieve high resolution soil moisture.

An integrated soil moisture retrieval algorithm is presented in this work wherein both radar and radiometer measurements are used simultaneously to retrieve surface soil moisture. This method is applied to bare rough surfaces with varying soil moisture and roughness distributions. A thresholding method based on physical models of scattering and emission is presented to obtain equivalent estimated brightness temperatures from active radar measurements. This technique is used as a constraint to link active and passive data within the optimization scheme. Numerical simulations are performed to investigate the proposed inversion technique using the method of Simulated Annealing.